Issue #5 · May 22, 2026

The Senior Analyst Brief Has Split. Most Résumés Address the Half That Isn't Hiring.

Career Snapshot

The Senior / Staff Data Analyst title is running two separate job briefs right now. The product analytics version has been the default career target for five years. The advertiser measurement version is the one with active headcount at social platforms — and it tests a different set of competencies.

TikTok (US advertising org) has open Senior and Staff Analyst roles across ads measurement and revenue analytics. The brief requires incrementality testing experience, attribution framework fluency, and advertiser-side data pipeline familiarity. The product analytics vocabulary gets you a screen. The measurement vocabulary gets you a hire.

LinkedIn is adding senior analysts in the advertising measurement function — B2B attribution methodology, first-party signal auditing, IO-model pipeline coverage. Most candidates applying from consumer platforms do not have a prepared answer for the attribution window overlap question. The interview finds out.

X (Twitter advertising org) is in an active rebuild of its ads measurement and performance analytics function. Senior and Staff Analyst comp across social platforms in this track: $155K–$215K base.

Hot this week:

Senior / Staff Data Analyst — advertiser measurement. The active search at social platforms. Distinct prep from the product analytics track.
Staff Data Analyst — product analytics. Slower opens. The Q2 AI feature review cycle has stalled headcount additions at several platforms while teams evaluate what they shipped in Q1.
BI / Analytics Leadership. Quiet, but Pinterest and Reddit have measurement build-outs underway that will produce Director openings in Q3.

Reading the tape: the product analytics brief has been the career-default for a decade. The advertiser measurement brief is the active hire. They do not share an interview rubric.

Skill Market Shifts

1. The Senior Analyst brief at social platforms has split into two structurally distinct tracks. Most résumés don’t distinguish between them.

The shift: Senior Analyst JDs at social platforms now route to either product analytics (funnel, retention, feature lift measurement) or advertiser measurement (incrementality testing, attribution window selection, ROAS frameworks, holdout methodology). The JDs look similar. The first-round questions do not.

The evidence: TikTok ads measurement, LinkedIn B2B attribution, and Pinterest performance analytics JDs all name advertiser measurement competencies — incrementality, geo holdout design, last-touch versus multi-touch attribution — that are absent from the same companies’ product analytics JDs in the same job family. The split is in the brief, not just the title.

The career translation: the advertiser measurement screen asks “describe your methodology for a geo-based incrementality test when user-level randomization isn’t available.” That is a specific question with a specific answer. Most candidates haven’t written it down before the call. The interview is the first time they think it through. That is not the right occasion to start.

2. The senior analyst screen has moved from “can you produce the number” to “can you explain why your number differs from finance’s.”

The shift: SQL fluency and BI tool proficiency are now assumed at the senior level. The front-pass screen appearing in senior analyst loops at LinkedIn, Reddit, and DoorDash in H1 2026 is different: can you diagnose a metric discrepancy, explain it to a non-technical stakeholder, and hold the position when they push back? Everyone has been in the meeting where two teams have different revenue numbers and the VP says “figure it out.” Almost no one has the document they wish they’d brought to that meeting.

The evidence: senior analyst loops at these companies added a variance-explanation exercise to the interview structure in early 2026 — present a discrepancy, walk the panel through the diagnosis and recommendation. The question is testing trust, not technique.

The career translation: write a variance explanation memo from your recent work. One metric where your number and the adjacent team’s differed. Three paragraphs: what each number was, why they diverged, which was right and why. The variance explanation exercise does not test whether you can build a model. It tests whether you’d tell a VP their number was wrong.

3. The senior analyst screen now includes a question the product analytics track didn’t historically ask: can you identify when the AI’s answer is wrong.

The shift: as Snowflake Cortex Analyst, Databricks AI/BI, and Power BI Copilot reach senior analyst workflows, a new screen appears in Staff-level loops — not “can you operate the tool” but “can you tell when it’s wrong.” The tool will produce an answer. It will not flag when that answer is incorrect. This has been mistaken, organizationally, for a product-limitation problem. It is an analyst-skill problem, and the brief is beginning to test for it.

The evidence: Staff Analyst JDs at companies with production AI query layers have added “validation of automated outputs” and “analytical judgment” language in H1 2026. The interview will ask for a demonstration, not a claim.

The career translation: the artifact this interview wants is a written audit of an AI-generated answer you caught being wrong. The lead is business consequence, not technique. “I caught this because if we’d used that number in the advertiser QBR, the client would have churned.” The interview is looking for the consequence — what would have broken, what was at stake, what changed because someone noticed.

Next Action Plan

Portfolio. Target: Senior / Staff Analyst — advertiser measurement track. Write one incrementality test design — one page. Hypothetical or real. Define: advertiser type, measurement objective (did the campaign cause the purchase?), holdout methodology, primary metric, minimum detectable effect, one named confound. The advertiser measurement interview is asking for this without naming it.

Networking. One message to a measurement or analytics lead at TikTok, LinkedIn, or Reddit’s advertising org. Three sentences: “How does your team handle the boundary between product analytics and advertiser measurement — separate IC tracks, or shared function with a methodology split?” That signals you understand the structural change happening at these orgs. “What’s the culture like?” does not.

7-day micro-skill. Write a variance explanation memo. One metric you own or recently owned. Three paragraphs: your number, the adjacent team’s number, the assumption that explains the gap — focused on the assumption, not the query difference. By Friday you have the artifact the senior analyst loop is increasingly requesting, and almost no one has prepared it in writing.

Sources

Crafting runs weekly. The job is to be useful, not to sell. If a line in here is close to a move you're working on, the email reaches a person.

Hiring rather than hunting? Signal is the letter for your side of the desk.

Get next week's issue Email me

← All Crafting issues · Drafted with Claude · Edited by Paul Brown